Skip to main content
Figure 8 | Journal of Cheminformatics

Figure 8

From: Prediction of the potency of mammalian cyclooxygenase inhibitors with ensemble proteochemometric modeling

Figure 8

Target-averaged model performance. The number of datapoints is displayed through the size of the squares. Targets annotated with less than 30 compounds or with chemical structures displaying high structural diversity (Oryctolagus cuniculus COX-1, Rattus norvegicus COX-1, Bos taurus COX-1, and Bos taurus) are produced with high mean RMSEtest values. These observations indicate that PCM models are not always able to extrapolate in the chemical or the target space if a given target or compound family is not sufficiently represented in the dataset.

Back to article page